Crash in loiter because no failsafe

ArduCopter Copter 3.2
1

this video shows a loiter flight with all the failsafe disabled and battery monitor cable disconnected.
I was focused on vibrations checking… and i forgot the rest…

the drone lost connection with ground station, then it continued its flight until battery end.
ended in a crash.

i can’t find anything interesting on the logs. apparently nothing abnormal was logged.
enjoy the movie story.
let me know if this is normal (i was expecting the drone to stand in a gps point, and not continuing to move)

[youtube]https://www.youtube.com/watch?v=6ymTBbhV7D0[/youtube]

note: vibrations are really good (+1 -1)

apm with ublox gps, low cost frame, 3d printed gimbal designed by me, legs, case for electronic, dampening system…everything designed by me.

2

You need to post a DF log from on-board the controller. Why didn’t you switch to stabilize and fly it back?

3

I imagine what happened was that because all the failsafes were off when it lost radio contact (for whatever reason) it simply kept using the last valid radio inputs it had. So maybe the pitch was forward and the throttle stick a bit above zero… it would just continue in that direction then (as it did).

All I can say is the failsafes are there for a reason. Don’t turn them off!

Pretty lucky that it came down in a bush like that.

I guess all is well that ends well.

4

[quote=“rmackay9”]I imagine what happened was that because all the failsafes were off when it lost radio contact (for whatever reason) it simply kept using the last valid radio inputs it had. So maybe the pitch was forward and the throttle stick a bit above zero… it would just continue in that direction then (as it did).

All I can say is the failsafes are there for a reason. Don’t turn them off!

Pretty lucky that it came down in a bush like that.

I guess all is well that ends well.[/quote]

you are right.
failsafe were disabled cause i had problems with high vibrations (solved on these last flight), i didn’t spent time to setup the voltage threshold for the battery failsafe, and then i disabled them all.
I forgot the fact that the failsave also acts when groundstation connection becomes weak.

actually i’m investigating on the jellow effect on the camera.
i spent one day in settin the PID of the alexmos but i can’t get good results, even if i apparently well balanced the camera.

5

I use turnigy 9x receiver. this receiver gives no PPM output if i power off the transmitter.
Actually on mission planner i can see that the APM doens’t recognize the absence of the PPM infact it continues to show the last throttle value, even if i powered it off.
i read about a ppm encoder firmware update but i can’t find a clear explanation, i wish to have confirmation that i need to do it, and what exacly i have to do to solve the problem.

6

i detailed my questions and my problem in a dedicated thread,
i hope that you can reply as soon as possible.
this is the link to the discussion: viewtopic.php?f=100&t=10167&p=24912#p24912

this thread will be closed assigning the correct answer to rmackay9.

7

I found the problem origin:

On this video is well explained that turnigy receiver channel 4 and 5 does not stop the PWM in case of radio connection lost: youtube.com/watch?v=I6lVYA7z9NE

this means that i have to use the other channels.
in my setup this caused me a lot of pain.

now the questions are the following:

  1. did you changed something in firmware updates? i can’t remember this problem before, but it is also possible that i changed the connections

  2. don’t you think that the firmware can be stronger if you activate the failsafe if you find at least 2 channels dead for at least 1 second?
    why you expect all the channels to be dead simultaneously?

8

ok, problem solved, it was a bug in the PPM ENCODER. check the linked discussion on my previous post to read the solution.